Add Model
guoqingbao/xinfer
Adapt and port new LLM model architectures to this xinfer project.
Set up the minimal set of artifacts (tokenized DCLM corpus shard + released HuggingFace checkpoint converted to DCP) required to benchmark, profile, or regression-test a MoE model in PithTrain.
Install Setup Benchmark Inputs
$ npx skills add mlc-ai/pith-train --skill setup-benchmark-inputs -a claude-codeProject install by default; add -g for ~/.claude/skills/.
$ gh skill install mlc-ai/pith-train setup-benchmark-inputs --agent claude-codeProject scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).
$ git clone --depth 1 https://github.com/mlc-ai/pith-train.git skills-src && mkdir -p .claude/skills && cp -r skills-src/.agents/skills/setup-benchmark-inputs .claude/skills/setup-benchmark-inputs && rm -rf skills-srcUse ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.
Claude Code skills documentation · loads skills from .claude/skills/
Install the "setup-benchmark-inputs" agent skill from https://github.com/mlc-ai/pith-train/tree/main/.agents/skills/setup-benchmark-inputs into .claude/skills/setup-benchmark-inputs/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "setup-benchmark-inputs", then confirm the skill loads.Claude Code copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$skill-installer install https://github.com/mlc-ai/pith-train/tree/main/.agents/skills/setup-benchmark-inputsType this inside Codex. $skill-installer <name> installs a curated skill from openai/skills. The installer writes to $CODEX_HOME/skills (default ~/.codex/skills). Restart Codex if the skill does not show up.
$ npx skills add mlc-ai/pith-train --skill setup-benchmark-inputs -a codexProject install goes to .agents/skills/; add -g for ~/.codex/skills/.
$ gh skill install mlc-ai/pith-train setup-benchmark-inputs --agent codexProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/mlc-ai/pith-train.git skills-src && mkdir -p .agents/skills && cp -r skills-src/.agents/skills/setup-benchmark-inputs .agents/skills/setup-benchmark-inputs && rm -rf skills-srcUse ~/.agents/skills/ instead of .agents/skills for a personal install.
Codex skills documentation · loads skills from .agents/skills/
Install the "setup-benchmark-inputs" agent skill from https://github.com/mlc-ai/pith-train/tree/main/.agents/skills/setup-benchmark-inputs into .agents/skills/setup-benchmark-inputs/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "setup-benchmark-inputs", then confirm the skill loads.Codex copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add mlc-ai/pith-train --skill setup-benchmark-inputs -a cursorProject install goes to .agents/skills/; add -g for ~/.cursor/skills/.
$ gh skill install mlc-ai/pith-train setup-benchmark-inputs --agent cursorProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/mlc-ai/pith-train.git skills-src && mkdir -p .cursor/skills && cp -r skills-src/.agents/skills/setup-benchmark-inputs .cursor/skills/setup-benchmark-inputs && rm -rf skills-srcUse ~/.cursor/skills/ instead of .cursor/skills for a personal install.
Cursor skills documentation · loads skills from .cursor/skills/, .agents/skills/, .claude/skills/, .codex/skills/
Install the "setup-benchmark-inputs" agent skill from https://github.com/mlc-ai/pith-train/tree/main/.agents/skills/setup-benchmark-inputs into .cursor/skills/setup-benchmark-inputs/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "setup-benchmark-inputs", then confirm the skill loads.Cursor copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ gemini skills install https://github.com/mlc-ai/pith-train.git --path .agents/skills/setup-benchmark-inputs--scope user (default) or --scope workspace; --path is the subfolder of the repo that holds the skill; --consent skips the security confirmation prompt.
$ npx skills add mlc-ai/pith-train --skill setup-benchmark-inputs -a gemini-cliProject install goes to .agents/skills/; add -g for ~/.gemini/skills/.
$ gh skill install mlc-ai/pith-train setup-benchmark-inputs --agent gemini-cliProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/mlc-ai/pith-train.git skills-src && mkdir -p .gemini/skills && cp -r skills-src/.agents/skills/setup-benchmark-inputs .gemini/skills/setup-benchmark-inputs && rm -rf skills-srcUse ~/.gemini/skills/ instead of .gemini/skills for a personal install, then run /skills reload.
Gemini CLI skills documentation · loads skills from .gemini/skills/, .agents/skills/
Install the "setup-benchmark-inputs" agent skill from https://github.com/mlc-ai/pith-train/tree/main/.agents/skills/setup-benchmark-inputs into .gemini/skills/setup-benchmark-inputs/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "setup-benchmark-inputs", then confirm the skill loads.Gemini CLI copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ gh skill install mlc-ai/pith-train setup-benchmark-inputsInstalls for Copilot at project scope by default; add --scope user for a personal install. Preview a skill first with gh skill preview. Needs GitHub CLI 2.90.0 or later (public preview).
$ npx skills add mlc-ai/pith-train --skill setup-benchmark-inputs -a github-copilotProject install goes to .agents/skills/; add -g for ~/.copilot/skills/.
$ git clone --depth 1 https://github.com/mlc-ai/pith-train.git skills-src && mkdir -p .github/skills && cp -r skills-src/.agents/skills/setup-benchmark-inputs .github/skills/setup-benchmark-inputs && rm -rf skills-srcUse ~/.copilot/skills/ instead of .github/skills for a personal install. Commit .github/skills so cloud agent and code review can use it.
GitHub Copilot skills documentation · loads skills from .github/skills/, .claude/skills/, .agents/skills/
Install the "setup-benchmark-inputs" agent skill from https://github.com/mlc-ai/pith-train/tree/main/.agents/skills/setup-benchmark-inputs into .github/skills/setup-benchmark-inputs/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "setup-benchmark-inputs", then confirm the skill loads.GitHub Copilot copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add mlc-ai/pith-train --skill setup-benchmark-inputs -a opencodeOpenCode documents no install command of its own. Project install goes to .agents/skills/; add -g for ~/.config/opencode/skills/.
$ gh skill install mlc-ai/pith-train setup-benchmark-inputs --agent opencodeProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/mlc-ai/pith-train.git skills-src && mkdir -p .opencode/skills && cp -r skills-src/.agents/skills/setup-benchmark-inputs .opencode/skills/setup-benchmark-inputs && rm -rf skills-srcUse ~/.config/opencode/skills/ instead of .opencode/skills for a personal install.
OpenCode skills documentation · loads skills from .opencode/skills/, .claude/skills/, .agents/skills/
Install the "setup-benchmark-inputs" agent skill from https://github.com/mlc-ai/pith-train/tree/main/.agents/skills/setup-benchmark-inputs into .opencode/skills/setup-benchmark-inputs/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "setup-benchmark-inputs", then confirm the skill loads.OpenCode copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
Facts
- Skill name
setup-benchmark-inputs- GitHub stars
- 355
- Token cost
- ~399 tokens
- SKILL.md length
- 59 words
- Files
- 3 (incl. scripts)
- Skills in repo
- 10
- Repo updated
- First seen
- Licence
- Apache-2.0
At a glance
Set up the minimal set of artifacts (tokenized DCLM corpus shard + released HuggingFace checkpoint converted to DCP) required to benchmark, profile, or regression-test a MoE model in PithTrain.
- The user asks to prepare benchmark inputs
- SKILL.md covers Prerequisites and Usage
- Runs Shell and Python scripts from its folder; calls bash
- Set up the benchmark workspace
What it does
Setup Benchmark Inputs is an agent skill from mlc-ai/pith-train. Set up the minimal set of artifacts (tokenized DCLM corpus shard + released HuggingFace checkpoint converted to DCP) required to benchmark, profile, or regression-test a MoE model in PithTrain. Use when the user asks to "prepare benchmark inputs", "set up the benchmark workspace", "download the DCLM shard", "fetch and convert the released checkpoint", "tokenize DCLM for DeepSeek/Qwen3", or when a downstream skill (capture-nsys-profile, validate-correctness, or any short canonical run) needs its workspace…
Its SKILL.md is about 400 tokens, which your agent loads only when the skill is triggered. The skill folder holds 3 other files, including scripts (for example `scripts/launch_setup.sh` and `scripts/setup.py`).
It sits in AI & LLM Engineering, covering Model hubs and datasets. It works with Hugging Face, DeepSeek and Qwen. The repository describes itself as: Compact and Agent-Native MoE Training System. The licence is Apache-2.0.
When your agent uses it
- The user asks to prepare benchmark inputs
- Set up the benchmark workspace
- Download the DCLM shard
- Fetch and convert the released checkpoint
Example prompts
- “prepare benchmark inputs”
- “set up the benchmark workspace”
- “download the DCLM shard”
- “/setup-benchmark-inputs”
Requirements
- Python 3
- A Bash shell
What it can do on your machine
Read from SKILL.md and the folder at commit c7c8b1d. It shows what the files ask for, not the result of running them.
Tool permissions
Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.
From allowed-tools in the SKILL.md frontmatter.
Runs code
Ships 2 files in scripts/ (Shell and Python), which the agent can run.
Shell commands in SKILL.md call:
bash
From the folder's file list and the shell code blocks in SKILL.md.
Network
No URLs in SKILL.md.
From URLs in SKILL.md, links to its own repository left out.
Credentials
Names no API keys, tokens, secrets or passwords.
From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.
Context cost
Setup Benchmark Inputs loads about 399 tokens when it runs. Until then it costs about 173 tokens; SKILL.md has 59 words of instructions outside code blocks.
Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.
Safety
Auto-check passed
The automated check found no risky patterns in SKILL.md.
Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); the scripts in this folder are not scanned.
SKILL.md
The full file from mlc-ai/pith-train at commit c7c8b1d, republished under its Apache-2.0 licence (© mlc-ai). 59 words, ~399 tokens.
Download SKILL.mdSave it as
.claude/skills/setup-benchmark-inputs/SKILL.md (or your agent's skills folder). This skill also uses 2 other files; get the full folder from GitHub.- name
- setup-benchmark-inputs
- description
- Set up the minimal set of artifacts (tokenized DCLM corpus shard + released HuggingFace checkpoint converted to DCP) required to benchmark, profile, or regression-test a MoE model in PithTrain. Use when the user asks to "prepare benchmark inputs", "set up the benchmark workspace", "download the DCLM shard", "fetch and convert the released checkpoint", "tokenize DCLM for DeepSeek/Qwen3", or when a downstream skill (capture-nsys-profile, validate-correctness, or any short canonical run) needs its workspace pre-populated. Produce `workspace/datasets/dclm-baseline/toktxt/<model>` and `workspace/checkpoints/<model>/torch-dcp/00000000`. Idempotent and safe to re-run.
Setup Benchmark Inputs
Setup the minimal artifacts needed to benchmark, profile, or regression-test a MoE model in PithTrain: a single DCLM corpus shard tokenized for the target model, and the released HuggingFace checkpoint converted to DCP format. Each step is idempotent (skips if its output already exists).
Prerequisites
- Python environment: activate
.venvin the repo root (source .venv/bin/activate).
Usage
mkdir -p logging
# Single-node (DeepSeek-V2-Lite)
bash .agents/skills/setup-benchmark-inputs/scripts/launch_setup.sh --model deepseek-v2-lite 2>&1 | tee logging/setup-deepseek-v2-lite.log
# Multi-node via SLURM (Qwen3-30B-A3B) — anchor the step with the launch-with-slurm skill
srun --jobid=<jobid> -W 0 -o logging/setup-qwen3-30b-a3b.log .agents/skills/setup-benchmark-inputs/scripts/launch_setup.sh --model qwen3-30b-a3b© mlc-ai, Apache-2.0. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file
Files
SKILL.md and 2 other files (scripts) in .agents/skills/setup-benchmark-inputs of mlc-ai/pith-train.
- SKILL.md
- scripts/launch_setup.sh
- scripts/setup.py
Open the folder on GitHubat commit c7c8b1d
Compare with similar skills
Setup Benchmark Inputs next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.
| Skill | Stars | Used in | Tokens | Auto-check | Licence | Repo updated |
|---|---|---|---|---|---|---|
| Setup Benchmark Inputs this skillmlc-ai/pith-train | 355 | — | ~399 | Automated safety check: Pass | Apache-2.0 | |
| Add Modelguoqingbao/xinfer | 333 | — | ~4.2k | Automated safety check: Notes | MIT | |
| Megatron Memory Estimatoryzlnew/infra-skills | 149 | — | ~2.2k | Automated safety check: Pass | None | |
| Qwen Mtp GgufR6410418/Jackrong-llm-finetuning-guide | 1.7k | — | ~1.7k | Automated safety check: Pass | MIT | |
| Veomni New ModelByteDance-Seed/VeOmni | 2.2k | — | ~2k | Automated safety check: Pass | Apache-2.0 | |
| Hf Architecture Tikzyzlnew/infra-skills | 149 | — | ~2k | Automated safety check: Pass | None |
Similar skills
Megatron Memory Estimator
yzlnew/infra-skills
Estimate GPU memory usage for Megatron-based MoE (Mixture of Experts) and dense models.
Qwen Mtp Gguf
R6410418/Jackrong-llm-finetuning-guide
Complete agent-ready workflow for Qwen-family MTP or nextn GGUF conversion and release.
Veomni New Model
ByteDance-Seed/VeOmni
A skill your agent uses when adding support for a new model to VeOmni.
Hf Architecture Tikz
yzlnew/infra-skills
Draw Sebastian-Raschka-gallery-style TikZ architecture diagrams for any HuggingFace decoder-only LLM, with per-block parameter formulas and concrete numbers.
Resolve
alexziskind1/model-shelf
Always resolve Hugging Face models via model-shelf before any download.
More from mlc-ai/pith-train
All 10 skills in this repoAnalyze Nsys Profile
mlc-ai/pith-train
Query a captured PithTrain Nsight Systems profile to measure compute/communication overlap, locate exposed comm by DualPipeV stage, and inspect per-rank stream behavior.
Capture Nsys Profile
mlc-ai/pith-train
Capture a Nsight Systems (.nsys-rep) profile of a short PithTrain run for performance analysis.
Validate Correctness
mlc-ai/pith-train
Validates that code changes do not break training correctness by comparing loss deltas against a base-vs-base run-to-run envelope.
Validate Performance
mlc-ai/pith-train
Measures the throughput difference between two branches with force-balanced routing.
Add New Model
mlc-ai/pith-train
Adds support for a new MoE language model to PithTrain. An agent skill from mlc-ai/pith-train.
Wandb Tracking
mlc-ai/pith-train
Read, analyze, and manage Weights & Biases (wandb) experiment data for PithTrain runs.
Related
Works with
Categories
Questions about Setup Benchmark Inputs
What does Setup Benchmark Inputs do?
Set up the minimal set of artifacts (tokenized DCLM corpus shard + released HuggingFace checkpoint converted to DCP) required to benchmark, profile, or regression-test a MoE model in PithTrain. Setup Benchmark Inputs is an agent skill from mlc-ai/pith-train. Set up the minimal set of artifacts (tokenized DCLM corpus shard + released HuggingFace checkpoint converted to DCP) required to benchmark, profile, or regression-test a MoE model in PithTrain.
When should I use Setup Benchmark Inputs?
Setup Benchmark Inputs fits situations like: the user asks to prepare benchmark inputs; set up the benchmark workspace; download the DCLM shard; fetch and convert the released checkpoint.
How do I install Setup Benchmark Inputs in Claude Code?
Run `npx skills add mlc-ai/pith-train --skill setup-benchmark-inputs -a claude-code`. Or copy the skill folder (.agents/skills/setup-benchmark-inputs in mlc-ai/pith-train) into .claude/skills/setup-benchmark-inputs in your project. Claude Code loads it when a task matches its description.
How do I install Setup Benchmark Inputs in Codex?
Run `npx skills add mlc-ai/pith-train --skill setup-benchmark-inputs -a codex`. Or copy the skill folder (.agents/skills/setup-benchmark-inputs in mlc-ai/pith-train) into .agents/skills/setup-benchmark-inputs in your project. Codex loads it when a task matches its description.
Can I use Setup Benchmark Inputs in Cursor, Gemini CLI or GitHub Copilot?
Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add mlc-ai/pith-train --skill setup-benchmark-inputs -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/setup-benchmark-inputs, .gemini/skills/setup-benchmark-inputs, .github/skills/setup-benchmark-inputs and .opencode/skills/setup-benchmark-inputs in your project.
What does Setup Benchmark Inputs need to run?
Going by SKILL.md and its folder, Setup Benchmark Inputs needs a shell and Python for the scripts in its folder and the command-line tools its instructions call (bash). Our summary lists: Python 3; A Bash shell.
Does Setup Benchmark Inputs access the network?
SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.
Is Setup Benchmark Inputs safe to install?
Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. The check reads SKILL.md only: the scripts in the folder are not scanned, so read them before running anything.
What licence does Setup Benchmark Inputs use?
Setup Benchmark Inputs is published under the Apache-2.0 licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.
How many tokens does Setup Benchmark Inputs use?
About 399 tokens (SKILL.md is roughly 1.6k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.
What are the alternatives to Setup Benchmark Inputs?
Skills that share tags, products or a category with Setup Benchmark Inputs: Add Model (guoqingbao/xinfer, 333 stars), Megatron Memory Estimator (yzlnew/infra-skills, 149 stars), Qwen Mtp Gguf (R6410418/Jackrong-llm-finetuning-guide, 1.7k stars) and Veomni New Model (ByteDance-Seed/VeOmni, 2.2k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.
Who maintains Setup Benchmark Inputs?
mlc-ai (a GitHub organization) maintains it in mlc-ai/pith-train, which has 355 GitHub stars. The repository holds 10 skills in this directory. The repository was last updated on October 4, 2026.
Source: mlc-ai/pith-train on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.